Reversible Colour Density Compression of Images using cGANs

TL;DR

This paper explores using cGANs to enable reversible, visually lossless colour density image compression, demonstrating its potential as an efficient compression method.

Contribution

It introduces a novel approach employing cGANs for reversible colour density image compression, improving upon traditional methods.

Findings

Produces visually lossless image generations

Demonstrates viability of efficient colour compression

Shows effectiveness of cGAN-based transformation

Abstract

Image compression using colour densities is historically impractical to decompress losslessly. We examine the use of conditional generative adversarial networks in making this transformation more feasible, through learning a mapping between the images and a loss function to train on. We show that this method is effective at producing visually lossless generations, indicating that efficient colour compression is viable.